The technical interview was a pretty absurd format. They give you a dataset and you have 50 minutes to understand it, analyze it, develop a model for it, and assess the model.

You do not have much time to do things properly, so it did not feel like an appropriate interview.

If you want to pass this interview, you will need to memorize the pandas and sci-kit learn APIs. While it is technically "open note," you cannot use a second monitor, and you will probably be judged for having your own notes.

The interviewer also does not provide much guidance on what is expected of you and will just let you fumble around the problem while remaining silent.

. . . . .